Retinal degeneration is a devastating affliction that leads to irreversible loss of vision, resulting from the progressive death of rod and cone photoreceptor cells in the outermost layer of the retina. One of the most promising avenues for replacing photoreceptors following degeneration is the introduction of photoreceptor precursor cells into the retina, which can differentiate into rods and cones. However, in trials of precursor transplant therapy, too many rods and not enough cones are produced. The result of these trials reveals a major hurdle for restoration of vision using stem-cell therapies, and highlights a criticl gap in our knowledge of photoreceptor formation: We do not know the mechanism that switches precursors from production of cone photoreceptors, which arise first, to production rod photoreceptors, which arise second. In order to understand how the number of cone versus rod photoreceptors is regulated, we have established a simple but powerful tool to study photoreceptor development, the zebrafish pineal organ, which shares a very close evolutionary relationship with the eyes. Just as in the retina, the zebrafish pineal contain fully developed rod and cone photoreceptors; in addition, the pineal photoreceptors are molecularly quite similar to those in the retina. A major advantage of the zebrafish pineal for studying photoreceptor development is its simplicity, which allows imaging of the live or fixed pineal organ easily in whole mount embryos without sectioning since there is no interference from the lens, aqueous humor, or overlying retinal layers. Recent studies from our laboratory indicate a novel and unexpected mechanism, called an incoherent feed-forward system, which controls the number of cone and rod cells produced by photoreceptor precursors. We propose that in early photoreceptor precursors, which give rise to cone cells, time-limited expression of the transcription factor Tbx2b expression drives the expression of cone-specific genes and also initiates expression of the cone-gene repressor Nr2e3. Nr2e3 protein continues to accumulate until it reaches a threshold concentration that shuts off cone gene expression and allows rods to form instead. In order to test the implications of our feed-forward hypothesis for photoreceptor formation, we will take advantage of the genetic tools that we and others have generated in zebrafish. Then we will leverage our knowledge to test the effects of a disease-causing mutation in Nr2e3 on feed-forward control. Our studies will illuminate the molecular control of cone versus rod production, which will provide invaluable information for laboratories developing retinal replacement therapies as well as insight into the ontogeny of inherited retinal degeneration disorders.